How India's Brahmi became Southeast Asia's alphabets (~200 BCE)

Southeast Asia before writing

In the last centuries before the Common Era, the lands between the Irrawaddy and the Java Sea were among the most technically accomplished societies on earth that did not write. In the Red River delta of what is now northern Vietnam, the Dong Son culture was casting bronze drums of up to seventy kilograms — instruments whose tympana carry concentric friezes of plumed dancers, boats of the dead, and deer — using lost-wax methods that required precise control of alloy ratios and pour temperatures ¹⁶. At Co Loa, near modern Hanoi, a rampart city enclosing some six hundred hectares had risen by the third century BCE, its earthworks among the largest in Asia ¹⁶. None of these societies kept a single written record.

The same pattern held across the mainland. At Ban Chiang and Ban Non Wat on the Khorat Plateau of northeastern Thailand, communities had been working bronze since the second millennium BCE and iron since roughly the fifth century BCE, burying their dead with painted pottery, bangles, and socketed tools in cemeteries that archaeologists have read, layer by layer, as records of rising social rank ¹⁶. Along the central Vietnamese coast, the Sa Huynh culture interred its dead in lidded ceramic jars with carnelian, agate, and the distinctive double-headed animal ear ornaments that travelled as far as the Philippines and Taiwan — evidence of sea-lane exchange networks centuries older than any Indian contact ⁸. These were stratified, metallurgically sophisticated, far-trading societies. What they did not have was script.

What an oral world carried in its head

The absence of writing was not an absence of knowledge. The societies of pre-literate Southeast Asia maintained, by memory and apprenticeship alone, bodies of expertise that still impress the specialists who reconstruct them:

• Navigation: Austronesian-speaking sailors had settled the islands from Sumatra to the Philippines millennia earlier, and their descendants ran regular exchange routes across the South China Sea by reading swells, stars, and bird flight ¹⁶.
• Metallurgy: the Dong Son drum casters and the bronze workers of the Khorat Plateau transmitted alloy recipes and mold-making techniques across generations without a written formula ¹⁶.
• Rice agronomy: wet-rice cultivation, with its calendars of flood and transplanting, ran on orally held seasonal knowledge.
• Genealogy and law: descent, marriage alliance, debt, and feud — the operating system of chiefly politics — were held in trained memory, recited and contested aloud.

Every one of these systems had a built-in constraint: knowledge died with its holders unless deliberately, laboriously passed on. There were no archives, no contracts, no king lists, no scripture. When the first outside observers arrived who could write, Southeast Asia entered the historical record through other people's eyes — and the first thing those observers noted was how much was already there.

Cities before letters

The scale of what already existed is easy to underestimate, because for a century the region's history was written backward from its Indian borrowings. Radiocarbon work of the past two decades has corrected the picture. At Sri Ksetra in the Irrawaddy basin of central Burma, charcoal from the great brick walls has returned dates between roughly 50 and 200 CE — meaning the Pyu were raising one of Southeast Asia's earliest cities, an enclosure whose walls run some thirteen kilometers around, at the very beginning of the contact era, not as its result ⁹. The Pyu sites — Beikthano, Halin, Sri Ksetra — show iron-working, irrigation tanks, and urban planning whose foundations owe nothing to the Ganges ⁹¹⁶.

The same lesson comes from the Mekong. Angkor Borei, the delta city that would later produce the oldest dated Khmer inscription, was a moated settlement with brick architecture and a dense ceramic sequence reaching back into the first millennium BCE ¹¹¹⁶. The canal system that linked it toward the coast at Oc Eo — sections of it traced by aerial photography across dozens of kilometers of delta — is engineering of a high order, and its builders were the descendants of the delta's own Iron Age communities, not Indian colonists ¹⁵¹⁶. When the letters arrived, in other words, they arrived in societies that already had cities to govern, surpluses to record, and elites with something to say about themselves. That is precisely why the letters took.

The monsoon corridor

Geography had long pointed the region toward India. The monsoon system of the Bay of Bengal is a seasonal conveyor: from roughly November to February the winds blow reliably from the northeast, carrying ships from the Ganges delta and the Coromandel coast toward the Thai-Malay peninsula; from May to September they reverse and carry the ships home ². A merchant could sail east with one season, trade through the inter-monsoon calm, and return with the next. The crossing did not require heroic navigation. It required patience and a cargo worth the wait.

By the fourth century BCE — a full half-millennium before the first surviving inscription — that cargo was moving. At Khao Sam Kaeo, a hilltop settlement above the Tha Taphao river on the upper Thai-Malay peninsula, excavations directed by Bérénice Bellina between 2005 and 2009 uncovered what she has called one of the earliest port-cities in Southeast Asia: a fortified, industrially organized town of the fourth to second centuries BCE where Indian carnelian and agate beads were not merely imported but manufactured on site, by Indian techniques, alongside jades worked in Taiwanese styles and bronzes with Vietnamese and Han affinities ⁸. Glass and stone ornament workshops at the site reproduce South Asian production sequences so faithfully that Bellina argues for the presence of resident South Asian craftsmen, not just their goods ⁸. At Ban Don Ta Phet in west-central Thailand, a cemetery of the fourth century BCE has yielded carnelian lion pendants and etched beads of Indian manufacture ⁸.

This matters for what came later because it establishes the order of operations. The first Indians in Southeast Asia were not missionaries carrying scripture. They were traders and artisans carrying beads, and the exchange relationship they built ran for some four centuries before anyone wrote anything down that survives. The script, when it came, travelled a road that commerce had already paved.

The transmission: merchants first, then brahmins

Writing arrived in Southeast Asia as part of a package that historians, following George Coedès, have called Indianization — a process whose mechanism has been argued over for a century, in part because it is genuinely strange. Between roughly the second and fifth centuries CE, the courts of the Mekong delta, the central Vietnamese coast, Borneo, Java, and the Irrawaddy basin began to present themselves in Indian terms: Sanskrit royal names, Hindu and Buddhist cults, Indian calendrical systems, and Brahmi-derived letters. Yet no Indian state conquered an inch of Southeast Asian ground. There is no evidence of Indian colonization fleets, no settler kingdoms, no tribute extracted back to the Ganges ²⁷.

Funan and the view from China

The earliest Southeast Asian polity visible in any historical record is the state the Chinese annals call Funan, centered on the Mekong delta from around the first century CE. Its port, excavated at Oc Eo in the delta west of modern Ho Chi Minh City, has produced Roman medallions of Antoninus Pius and Marcus Aurelius, Indian intaglios, and Han bronze mirrors — a junction box of the trans-Asian sea trade ²¹⁵. In the 240s CE the Wu emperor sent two envoys, Kang Tai and Zhu Ying, to this kingdom; their reports, preserved in fragments in later Chinese compilations, are the first eyewitness description of any Southeast Asian society ¹².

What the envoys describe is a court in mid-transformation. The Liang shu, drawing on their account, reports of Funan: "They have books and depositories of archives and other things. Their characters for writing resemble those of the Hu" — the Hu being the Central Asian peoples whose scripts, like Brahmi's descendants, ran in Indic letterforms ¹²². The line is easy to read past and worth stopping on. Within a few generations of the first surviving local inscriptions, a Mekong delta kingdom was running written archives — and a Chinese observer, from the world's other great scribal civilization, identified the script at a glance as Indian-derived. By the third century CE, the technology of writing had not merely arrived in Southeast Asia; it had been institutionalized.

The same Chinese sources preserve Funan's own myth of its founding: a brahmin named Kaundinya arrived by ship, was met in arms by the local queen Liu-ye ("Willow Leaf"), defeated or married her — the versions vary — and their union founded the dynasty ²¹². The story is legend, not record, but it is the legend the Funanese elite themselves chose to tell, and historians from Coedès onward have read it as a compressed memory of how Indianization actually worked: foreign ritual specialists marrying into, and being absorbed by, a local power structure that remained sovereign ²⁶.

Suvarnabhumi: why the ships kept coming

Indian literature of the late centuries BCE already knew the lands across the bay by a name that explains the traffic: Suvarnabhumi, "the land of gold." The Jataka tales — the Buddhist birth-stories, parts of which were circulating by the last centuries BCE — send merchants sailing east to Suvarnabhumi as a stock plot device, the way later European stories sent younger sons to the Indies ². Behind the topos stood real cargoes. Mainland Southeast Asia held alluvial gold, and beyond it tin from the peninsula, aromatics, camphor, and the forest products — resins, rhinoceros horn, kingfisher feathers — that fed both Indian and Chinese demand ²⁸.

The traffic intensified in the first two centuries CE, and Coedès connected the surge plausibly to events far to the west: Roman appetite for eastern luxuries was draining gold out of the Mediterranean world, Indian merchants were in the middle of the exchange, and the sea road east was the supply side of the boom ². The Antonine medallions in the soil of Oc Eo — one struck for Antoninus Pius in 152 CE — are the physical trace of a single commercial system that ran, port to port, from the Red Sea to the Mekong delta ²¹⁵. Writing rode this system the way software rides hardware. Every durable trading post needed contracts, tallies, and letters; every resident Indian merchant community brought its literate specialists; and the local rulers who taxed the trade watched, for generations, what those marks on palm leaf could do — hold a debt steady across monsoon seasons, make an agreement outlive the men who made it ²⁸. The kings adopted the letters' prestige register later; the wharves had been demonstrating the letters' use all along.

Who carried the letters

Coedès, whose Les états hindouisés d'Indochine et d'Indonésie (1944, revised through 1964) built the field, defined Indianization as "the expansion of an organized culture that was founded upon the Indian conception of royalty, was characterized by Hinduist or Buddhist cults, the mythology of the Puranas, and the observance of the Dharmasastras, and expressed itself in the Sanskrit language" ¹². Note what the definition makes central: not population movement, not conquest, but a portable cultural operating system — and its scripting language.

Who physically carried it? The scholarship has converged on a layered answer ²⁶⁷:

1. Merchants opened and maintained the routes, from the fourth century BCE onward, but merchants alone do not teach a court Sanskrit ⁸.
2. Brahmins and Buddhist monks travelled the same ships from around the turn of the era — ritual specialists whom Southeast Asian rulers recruited, in I. W. Mabbett's reading, much as they recruited any other prestige craft ⁷.
3. Local rulers did the actual adopting. O. W. Wolters argued that Indian forms were "localized" — selectively taken up, restated, and bent to indigenous purposes by chiefs who found in Sanskrit kingship a technology for converting personal charisma into durable, inheritable authority ⁶.
4. Return travellers: Southeast Asians themselves sailed west. Chinese pilgrims' itineraries and the archaeology of the peninsula make clear the exchange ran in both directions, and some of the carrying was done by Southeast Asian ships ²⁸.

The older colonial-era picture — Indian colonists civilizing a passive periphery, a notion cultivated with particular enthusiasm by the "Greater India" school of the 1920s and 1930s — has not survived scrutiny. Mabbett's two 1977 essays dismantled the evidentiary basis for settler colonization; Wolters and his students rebuilt the story around Southeast Asian agency ⁶⁷. The current consensus is almost the inverse of the colonial one: Indianization happened because Southeast Asian elites pulled, not because Indian elites pushed. That is also why its costs were so unlike those of most transmissions this atlas records.

A script built to travel

The writing system the brahmins and monks carried was supremely portable. Brahmi — first securely attested in the rock and pillar edicts of the Mauryan emperor Ashoka, circa 250 BCE, and the ancestor of nearly every script from Tibet to Bali — is an abugida: each consonant letter carries an inherent vowel, modified by satellite marks ³¹⁴. Richard Salomon's standard survey of Indian epigraphy traces how the script's southern varieties, above all the Pallava script of the southeastern Indian coast, became the export models: Pallava-type letters of the fourth to sixth centuries CE are the direct templates for the earliest scripts of Cambodia, the Malay world, and Java ³⁴.

The abugida structure mattered for what happened next. Because the system encodes syllables compositionally, it could be retuned to languages phonologically remote from Sanskrit — Austroasiatic Khmer, Austronesian Malay and Cham, Tibeto-Burman Pyu and Burmese, and eventually Tai — by adding, dropping, and repurposing signs ³⁴. The Greeks had to hack vowels into Phoenician consonant letters; the Khmers and Javanese received a system whose architecture already anticipated adaptation. J. G. de Casparis, surveying a millennium of Indonesian palaeography, emphasized how quickly the imported letterforms began evolving locally — within two centuries, island Southeast Asia was developing script varieties with no exact Indian counterpart ⁴.

One material fact shapes everything we know about this story: the tropics destroy writing. The everyday media of early Southeast Asian literacy — palm leaf, bark paper, lacquered boards — rot in a monsoon climate within decades unless continuously recopied ³⁴. What survives from the first millennium is therefore almost exclusively what was cut into stone or stamped into metal, the formats reserved for gods and kings. The corpus is not a sample of what was written; it is a sample of what was meant to be eternal. At Sri Ksetra, the exception proves the rule in precious metal: the Khin Ba mound, excavated in 1926–27, yielded a manuscript of twenty gold leaves bound with gold wire, inscribed in the fifth or sixth century CE with excerpts from the Pali canon — among the oldest physically surviving Pali texts anywhere on earth, older than anything preserved in Sri Lanka, the tradition's homeland ⁹. Buddhist Burma was writing scripture for permanence within a century or two of learning to write at all. The letters' everyday work — the contracts, the tallies, the letters actually sent — has vanished, and the loss biases every generalization that follows. We see the cathedral uses of literacy; the market-stall uses we must infer ³⁹.

Ms Sarah Welch. Yupa pillar of King Mulavarman with Sanskrit inscription in early Pallava script, c. 400 CE. National Museum of Indonesia, Jakarta. CC0 via Wikimedia Commons. · CC0

What changed and what was replaced

Stone speaks Sanskrit first

The earliest surviving Southeast Asian writing is not in any Southeast Asian language. The Vo Canh stele, found in 1885 near Nha Trang on the central Vietnamese coast and now in the National Museum of Vietnamese History in Hanoi, carries a Sanskrit inscription in verse celebrating a ruler remembered by the name Sri Mara; Coedès placed it in the second or third century CE and read it as the oldest Sanskrit text in Southeast Asia, though paleographers including D. C. Sircar and, more recently, Anton Zakharov have argued for a fourth- or even fifth-century date ²¹⁰. Around 400 CE, on the Mahakam river in eastern Borneo — about as far from India as the Indianized world ever reached — King Mulavarman of Kutai raised seven stone sacrificial posts, yupas, inscribed in correct Sanskrit verse and early Pallava letters, commemorating his gifts to brahmins: thousands of cattle, quantities of gold ⁴¹⁵. His grandfather's name, Kundungga, is not Sanskrit but local — the dynasty had Indianized within living memory ².

Sheldon Pollock has given this phenomenon its sharpest framing: from roughly 300 to 1300 CE, an enormous arc of Asia from Afghanistan to Java constituted a "Sanskrit cosmopolis," in which courts that shared no political structure, no ethnicity, and no vernacular used a single prestige language for the public poetry of power ⁵. A king on the Mahakam announced himself in the same meters, the same gods, and the same letters as a king on the Ganges — not because either ruled the other, but because Sanskrit had become the medium in which rule itself was made legible ⁵. Writing arrived in Southeast Asia not as a bookkeeping convenience but as a component of this technology of majesty: the first thing the region's kings did with letters was praise gods and themselves, in someone else's language, for eternity ⁵¹⁵.

The vernacular turn

Then, within a few generations, the borrowed letters began to speak local languages — and this, more than the Sanskrit overlay, is the transformation that endures. The earliest secure dates ³⁴⁹¹⁰:

The Dong Yen Chau inscription, a short curse-formula protecting a sacred naga spring, is the oldest surviving text in any Austronesian language — the language family that stretches from Madagascar to Easter Island enters written history in central Vietnam, in Indian letters ²⁴. The Kedukan Bukit stone, cut at Palembang in 683 CE and now in the National Museum of Indonesia, is the oldest text in Malay: a king of rising Srivijaya, Dapunta Hyang, records a sacred journey, an army of twenty thousand, and the founding of a settlement ¹³. Coedès, who edited the stone in 1930, used it and its companions to conjure the entire Srivijayan thalassocracy — a maritime empire whose existence had been forgotten for centuries — back into history ¹³. In central Burma, the Pyu cities had been writing their Tibeto-Burman language in southern Brahmi-derived letters since perhaps the fifth century CE; the corpus assembled by Arlo Griffiths, Bob Hudson, Marc Miyake, and Julian Wheatley in 2017 — 184 inscribed objects, from gold-leaf Pali texts to mortuary urns — is the documentary residue of Southeast Asia's first literate urban civilization ⁹.

Each vernacular debut follows the same script, so to speak: the language appears first in the service of religion and royal property — curses, donations, dedications — wrapped in Sanskrit honorifics, using letterforms one or two steps removed from a south Indian model ³⁴. Writing percolated downward from the gods.

One parent, many children

From these beginnings descends nearly every traditional script between the Irrawaddy and the Pacific. The genealogy, compressed ³⁴⁹:

• Pyu and Mon letters at Bagan fused into the Burmese script — the Myazedi pillar of 1113 CE, with parallel texts in Pyu, Mon, Pali, and Burmese, catches the handover in a single stone ⁹.
• Old Khmer script, continuous from 611 CE, became the modern Khmer script — and its cursive forms were adapted, in the thirteenth century, into Thai (traditionally credited to King Ram Khamhaeng of Sukhothai, c. 1283, though the authenticity of the famous inscription bearing his name is a genuine scholarly controversy) and subsequently Lao ³.
• Pallava letters in the archipelago became Kawi, the script of Old Javanese court literature, which in turn fathered the Javanese hanacaraka, Balinese, and Sundanese scripts, the Bugis-Makassar lontara of Sulawesi, the Batak and Rejang scripts of Sumatra, and — carried island to island — the baybayin family of the Philippines ⁴.
• Cham script continues among Cham communities in Vietnam and Cambodia: the living descendant of the region's oldest vernacular literacy ⁴.

Today the national scripts of Myanmar, Thailand, Laos, and Cambodia — the daily writing of well over a hundred million people — are Brahmi's great-grandchildren, as are the ceremonial scripts of Java and Bali. Only Vietnam, which took its writing from China and later from Rome, and the Islamized and colonized zones that adopted Arabic-derived Jawi and Latin letters, stand outside the family ²⁴.

Scripture, statecraft, literature

Once the vernaculars had letters, the letters reorganized three domains in succession. Scripture came first. In the Irrawaddy and Chao Phraya basins, the script arrived already married to Pali, the canonical language of Theravada Buddhism; the Pyu gold leaves and the Mon inscriptions of Dvaravati carry canonical excerpts centuries before any local chronicle exists ⁹. When Bagan adopted Theravada as state religion in the eleventh century, the script-religion package became the durable cultural architecture of the western mainland: to this day, Burmese, Thai, and Lao children traditionally learned their letters in monastery schools, from monks, out of religious texts — a pedagogical circuit running directly back to the transmission ²⁹.

Statecraft followed. The inscriptions show Indic administrative technology being absorbed item by item: the Saka era calendar (the Kedukan Bukit stone opens by dating itself to Saka 605 — the scribe of Palembang counting years from a king's era in western India) ¹³; land-grant formulas modeled on Indian charters; legal digests in the Dharmasastra tradition that evolved into the dhammasattha codes of Burma and the thammasat of Siam, which remained the framework of law into the nineteenth century ²⁶. None of this made Southeast Asian states Indian — Wolters's point stands — but it gave them shared instruments: a way to date, to deed, to codify, mutually legible from Burma to Bali ⁶.

Literature came last and went furthest. By the ninth century, Old Javanese poets were composing kakawin — court epics in Indian meters — including a Ramayana that is not a translation but a re-imagining, its hero localized into a Javanese moral landscape ⁴⁵. Pollock treats the kakawin tradition as the signature achievement of the cosmopolis's vernacular turn: a literature that used Sanskrit's full poetic apparatus to say things Sanskrit had never said, in a language Sanskrit's makers had never heard ⁵. The pattern repeated across the region — Khmer, Mon, Burmese, Thai literary traditions each began as religious and royal writing and grew secular legs. Every one of them is a child of the borrowed letters.

What was displaced

Here the record must be honest about an absence. Nowhere in Southeast Asia is there secure evidence of writing before the Indian letters came. The transmission displaced no indigenous script — claims of pre-Indic writing systems, including proposed precursors of the Philippine baybayin, have no accepted epigraphic support ³⁴. What the letters displaced was subtler: the institutions of the oral world. The trained memory of the genealogist, the chanted boundary of the rice terrace, the recited claim to chiefly descent — wherever writing took hold, these lost their monopoly on permanence. A lineage recited can be renegotiated; a lineage inscribed on stone, in the language of the gods, cannot. Writing froze advantages that orality had kept fluid ⁵⁶.

And the freezing was selective. The letters came bundled with the rest of the Indic package: Shaiva and Buddhist cults installed above (never quite replacing) local ancestor and spirit worship; Sanskrit court titles layered over indigenous rank; the Indian ritual calendar; Dharmasastra legal concepts; and a varna vocabulary of brahmins and kshatriyas that Southeast Asian societies adopted in name while largely declining its substance — caste, as India knew it, never took root east of the Bay of Bengal ²⁶⁷. The landscape itself was renamed in Sanskrit: Suvarnabhumi, Dvaravati, Sri Ksetra, Ayutthaya — a toponymy of somewhere else, laid over the rivers and plains ². Localization meant Southeast Asians chose what to take. It did not mean the taking changed nothing.

What the cost was

The first dated sentence in Khmer is a list of human property

Set the scene precisely, because the document deserves it. Angkor Borei, in the Mekong delta of southern Cambodia: a moated city that had been one of Funan's centers. The date inscribed corresponds to 611 CE. The stone, catalogued by the French school as K. 557/600 and first edited by Coedès in 1942, is the earliest dated text in the Khmer language — the oldest dated vernacular document in mainland Southeast Asia ¹⁰¹¹. It is not a poem, not a chronicle, not scripture. It is an inventory of a temple's endowment: by Anton Zakharov's 2019 English translation, the donations include named dancing girls and singers, fifty-seven slaves — khnyum in the Old Khmer — along with cattle, rice fields, and plantations, conveyed to a god ¹⁰.

The detail that stings is the names. Coedès, in 1942, did not bother to translate the slave lists; Zakharov's edition restores them — men and women with Khmer, Sanskrit, Austronesian, and Austroasiatic names, each recorded precisely so that their obligation, and their children's, would outlast every living memory of it ¹⁰. Michael Vickery's survey of the seventh- and eighth-century corpus shows K. 557 is no anomaly: the pre-Angkorian inscriptions are dominated by exactly such transfers, hundreds of stones conveying thousands of unfree workers — field hands, weavers, musicians — to temple estates ¹¹. This is what the new technology was first used for in the vernacular. Not because writing created servitude: the chiefly societies of the pre-literate mainland assuredly knew bondage. But writing industrialized its bookkeeping. An obligation held in memory dies, fades, can be contested; an obligation cut into stone in the presence of a god is designed to be permanent ¹¹. The earliest function of vernacular literacy in Southeast Asia was to make hierarchy durable.

The Sanskrit ceiling

The second cost was exclusion by design. For the first four to five centuries of Southeast Asian literacy, the prestige register — eulogy, theology, law, diplomacy — was conducted in Sanskrit, a language no Southeast Asian population spoke and only a court-trained sliver could read ⁵. Pollock's "cosmopolis" was, seen from below, a ceiling: the vernaculars of perhaps the entire population were judged, for centuries, unfit for anything beyond property lists — Old Khmer appears in the inscriptions chiefly to itemize the slaves, fields, and cattle whose donation the Sanskrit portion celebrates in verse ⁵¹¹. Literacy itself remained a court and temple monopoly throughout the first millennium; there is no evidence of merchant, artisan, or agrarian writing in the entire early corpus ³⁴. The letters that would one day write Thai novels and Burmese newspapers spent their first five hundred years in Southeast Asia as an instrument of two institutions: the palace and the god-estate.

There is also a historiographical cost, paid much later, that belongs in this accounting because the script's foreign origin was its instrument. From the 1920s, the "Greater India" school of Indian nationalist historiography read the Brahmi-derived letters, the Sanskrit inscriptions, and the Indic temple architecture of Southeast Asia as proof of ancient Indian colonization — Hindu colonies, civilizing a barbarous periphery ⁷. Colonial French scholarship, Coedès included, framed the region as "hindouisé," Further India, a cultural annex ¹⁷. For half a century, Southeast Asia's own past was narrated as the achievement of someone else, and the region's pre-Indian accomplishments — the bronze, the cities, the seafaring — were barely looked for. The corrective scholarship of Mabbett, Wolters, and the archaeologists who dated Ban Chiang and Sri Ksetra had to dismantle a story that the inscriptions' own letterforms had seemed to underwrite ⁶⁷⁹. A borrowed script, read carelessly, became evidence against its borrowers' agency. The stones never said that; their readers did.

Dead languages in the family plot

Third, the genealogy of scripts has casualties in it. Pyu — the language of Southeast Asia's first literate urban culture, written for perhaps seven centuries — declined after the Burmese kingdom of Bagan absorbed the Pyu world; the Myazedi pillar of 1113, which gives Pyu one face of four beside the ascendant Burmese, is simultaneously the script's monument and, within a century or so, its tombstone: the language and its letters were extinct by about the thirteenth century ⁹. Mon, the prestige vehicle through which Indic letters and Theravada Buddhism reached the Burmese, was reduced over centuries of Burman conquest from imperial language to embattled minority tongue ²⁹. Cham literacy survived, but as the inheritance of a people whose kingdoms were progressively destroyed by the Vietnamese advance southward — Vijaya fell in 1471 with, the Vietnamese annals record, some sixty thousand killed and thirty thousand carried off ². These later destructions were the work of regional empires, not of the transmission itself; but they mean the family tree of Brahmi's Southeast Asian descendants is also a record of which courts crushed which.

Hybernator. Myazedi inscription, Pyu-language face, 1113 CE. Myazedi pagoda, Bagan. Public domain via Wikimedia Commons. · Public domain

Weighing the bill

Against the scale this atlas uses, the Brahmi transmission sits near the bottom of the cost range, and the reasons are instructive. The transfer was unaccompanied by conquest: no Indian army, no settler displacement, no extraction running back to the source ²⁷. The receiving elites were the initiators; Wolters's localization is, among other things, a description of consent ⁶. The deaths in this story — the temple slaves of the Khmer stones, the war captives of later script-bearing empires — were inflicted by Southeast Asian institutions on Southeast Asian people, with writing as the instrument of record rather than the cause ¹⁰¹¹.

But an instrument of record is not nothing, and a cost of zero would be false. The honest bill reads:

• Bondage made permanent: the earliest vernacular documents of the Khmer world are slave registers, and the temple-estate economy they record ran on inscribed, hereditary obligation ¹⁰¹¹.
• A five-century ceiling of sacred foreign language above every local tongue ⁵.
• Oral institutions stripped of authority wherever stone and palm-leaf reached ⁶.
• Languages buried in the family plot — Pyu extinct, Mon and Cham reduced — though by later empires wielding the scripts, not by the script's arrival ²⁹.

Five judgments, one record: the transmission of Brahmi to Southeast Asia is the rare entry in this atlas where the receiving cultures got one of the most consequential technologies in human history — their own written word, and with it their own recorded past — at a price paid almost entirely in coin they chose, eventually, to mint themselves. The alphabet that crossed the monsoon sea belonged, within three centuries, to the people it reached: bent to their languages, evolved past its models, and kept for fifteen hundred years and counting. What it cost them was the particular permanence it gave to their own hierarchies — the slave lists of Angkor Borei are legible today precisely because the tool that recorded them was built to defeat forgetting. That is the alphabet's double edge everywhere it has ever travelled: it remembers what the powerful wanted remembered, and it cannot stop remembering who paid.